Re: connection speed when using pon

From: Gene Heskett (gene.heskett_at_verizon.net)
Date: 01/31/05

  • Next message: Kevin Jones: "Unable to resolve hostnames with manual IP" 
    To: debian-user@lists.debian.org
Date: Sun, 30 Jan 2005 22:21:57 -0500

    On Sunday 30 January 2005 21:22, John Hasler wrote:
    >Gene writes:
    >> No, these are meaningfull numbers alright, and thats the actual
    >> speed the data is moving.
    >
    >But they do not take into consideration the compression done by the
    > modem.

    Oh, but it does. If it weren't for the compression used in the
    modems, the working baud rate would be limited to 1/4 or less of
    whats achieved. Both v42.bis and MN5 play a role in this, a very
    heavy role. Even at 9600 baud, v42.bis is needed just to get that.
    The rest of it is of course in the odd phase shifted keying as a way
    to pack more than one bit per cycle of the audio tone that is
    actually sent down the wire. That phase shift keying is the major
    reason you don't hear the individual tones of the code anymore, but a
    quite raucous, nearly white noise occupying the whole 300-2500 hz
    bandwidth of a voice grade line, if the line is listened in on.

    Some bandwidth is of course used in error corrections, and when the
    error corrections begin to impinge on the data rate, thats when the
    modems will call a temporary halt, tell us users hey we're busy hold
    up a few seconds, while they get together and negotiate a new, more
    robust but slower modulation method. Conversely, if the error rate
    drops to zilch for a while, they'll stop, and renegotiate a faster
    modulation method, up to the theoretical maximum of 57k in the
    download direction on a really good line.

    Much of this error correction is also because modern modems are
    capable of compareing what they send to whats comeing back slightly
    later in the form of echos, and adjusting their internal digital
    filter taps to cancel the echos that inevitably come back from poor
    junctions and terminations in every box that cable goes thru between
    you and the CO. Once at the CO, then its generally into radios or
    microwave stuff and in pretty good shape, but that 'last mile' of
    copper is, figuratively speaking, pure hell for a signal as complex
    as these modems can generate. Its got to be a lot better than what
    it takes for you to recognize your friends voice for a modem to
    function correctly.

    Todays 56k dual mode modem that runs on a couple of watts from a wall
    wart, has more computing power in it than your dads whole computer
    had 15 years (or even less) ago.

    >I suggest that the OP look at the pppstatus package.
    >--
    >John Hasler 

    -- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
99.32% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attorneys please note, additions to this message
by Gene Heskett are:
Copyright 2005 by Maurice Eugene Heskett, all rights reserved.
-- 
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
  • Next message: Kevin Jones: "Unable to resolve hostnames with manual IP"

    Relevant Pages

    • Re: SDSL router recommendations
      ... >Thats may turn out to be a bad choice in a changing world. ... >Buy modems and routers separately ... Tiscali - dialup speeds at Broadband prices, ...
      (uk.telecom.broadband)
    • Re: connection speed when using pon
      ... >Gene Heskett wrote: ... >> connection between the modems, it can be as high as 230kilobaud, ... No, these are meaningfull numbers alright, and thats the actual speed ...
      (Debian-User)